weights of silver, antimony, and sulphur, as determined by Dr. Thomson, for the more complex numbers employed by Bonsdorf in the preceding calculations.

Red silver ore is a compound of 1 integrant particle of sulphuret of silver and 1 integrant particle of sulphuret of antimony. Sulphuret of silver is composed of

The numbers in the last column are exceedingly near the result obtained by Bonsdorf. Indeed, if the loss in his analysis was sulphuret of antimony, as is exceedingly likely, the theoretic numbers almost coincide with those derived from the analytical experiments.

ARTICLE VI.

On the Characters of some Mineral Substances before the Blowpipe. By J. G. Children, FRS. &c.

THE blowpipe, when skilfully handled, is the most convenient chemical instrument for mineralogical researches on a small scale that has hitherto been invented. By its means we are enabled in a few minutes to determine the principal ingredients in any mineral submitted to our examination, even though it be composed of several elements. By merely directing the flame of a small lamp on a fragment about half the size of a peppercorn, supported on a piece of charcoal, or in the platina forceps, most of the volatile substances, as sulphur, arsenic, zinc, cadmium, antimony, bismuth, and tellurium, may be detected;

baryta will be known by the greenish-yellow, and strontita by the crimson colour it imparts to the flame. By employing only three fluxes, carbonate of soda, borax, and the triple phosphate of soda and ammonia (salt of phosphorus), with the occasional use of the nitrate of cobalt, we can readily ascertain the presence of silica, alumina, magnesia, and almost all the fixed metallic oxides; and by the further examination of the fused globule, especially that with carbonate of soda, by dissolving it in a drop of muriatic or nitric acid, on a slip of glass, and applying the proper tests, unequivocal evidence may be obtained of the presence of any of the other earths or oxides of which the substance is composed, and even a tolerable estimate_may_frequently be formed of their respective proportions. By substituting nitrate of baryta as the flux, and using a slip of platina foil for the support, instead of the wire, the presence of either of the alkalies may, by the usual well-known processes, be detected, with equal ease and certainty, on the same minute scale of operation.

An advantage peculiar to this microscopic chemistry is the very small quantity of matter that is sufficient for examination, which may generally be detached from rare and costly specimens without injury, whereas for operations on a larger scale, it is necessary wholly or in great measure to destroy them. When the exact proportions of the ingredients of a mineral are required, recourse must necessarily be had to more elaborate processes, but even then previous examination by the blowpipe is of essential service, since by indicating the different substances present, it enables us to determine the most advantageous method to be adopted in the subsequent analysis. Convinced of the utility of this sort of investigation, I propose, from time to time, to publish in the Annals the blowpipe characters of such minerals as have not already been so examined. For those which form the subject of the present communication, I am indebted to the kindness of Mr. Brooke.

1. Arfwedsonite. (Phillips's Mineralogy, p. 377.)

Alone in the glass matrass, gives off a very little moisture at a red heat: no decrepitation; appearance of the assay scarcely at all altered.

Alone in platina forceps, swells up, and fuses with great ease into a brilliant, opaque, black globule.

With soda, on platina wire, in the oxidating flame, fuses readily into a dark-brown opaque globule, while hot; olivegreen, cold. By the addition of nitre the green colour becomes much brighter. In the reducing flame the colour changes to a dark, slightly greenish-brown.

With borax, dissolves readily, and gives a transparent globule of a garnet-red colour, hot, which changes to a deep wine-yellow

on cooling. In the reducing flame, the colour is a deep bottlegreen.

With salt of phosphorus, the action is very slow and imperfect; the globule is transparent, and, while hot, has a deep wineyellow colour; when cold, it is colourless. In the reducing flame the colour is lighter, and more inclined to green, while hot; when cold, colourless. A considerable portion of the assay remains undissolved, in the form of a dark-grey silica skeleton.

2. Latrobite. (Phillips's Mineralogy, p. 380.)

Alone in the glass matrass, at a red heat, gives off pure water; no decrepitation.

Alone in forceps, fuses easily into a white enamel.

With soda, fuses into a semi-transparent, irregular globule, of a light azure colour, when cold. The colour of the globule is not uniform, spots of a deeper colour than the rest appearing scattered over the surface. By the addition of nitre, the blue colour is at first much exalted, and assumes a very slight greenish hue; but by long continued flaming, the blue colour disappears, and is succeeded by a peach-blossom red colour, very similar to that of the mineral. In the reducing flame, the colour is wholly destroyed.

With borax, dissolves very slowly, into a perfectly transparent, very light amethyst-coloured globule. In the reducing flame, the colour entirely disappears.

With salt of phosphorus, action slow, and solution imperfect; globule transparent, very light-yellow, hot; colourless and slightly opaline, cold. In the reducing flame, the globule is colourless and transparent, both hot and cold. An undissolved silica skeleton remains in the globule.

With nitrate of cobalt, the assay gives a fine blue colour, intensely deep on the fused edges.

By dissolving the soda globule in muriatic acid, &c. I obtained silica, alumina, lime, iron, and manganese.

The latrobite is accompanied by a dark-coloured, nearly black substance, dispersed through it, here and there, in minute specks, which have an uneven, shining fracture, but are too small to allow me to distinguish any thing more of their external characters.

With salt of phosphorus, they presented before the oxidating flame the phenomena detailed above, but in the reducing flame, they gave a transparent glass, colourless while hot, and of a fine, rather deep-amethyst colour when cold. This colour flies instantly on the globule being heated, and on its cooling to a certain point, returns as instantaneously. These dark specks, therefore, appear to be an ore of titanium.

I examined the mica, which is another concomitant of latro

bite, thinking it possible that it also might contain titanium; but it gave no indication of that metal, either when fused in the reducing flame, with salt of phosphorus alone, or with the addition of a small morsel of tin-foil.

3. The Matrix, or greyish-coloured substance, in which the latrobite is imbedded.

Alone in the matrass, behaves like latrobite; appearance unaltered.

In forceps, bubbles up, and fuses into an irregular greyish globule.

With soda, in proper proportion to the assay, fuses into a greenish-grey, semi-transparent globule, which in the reducing flame is colourless. On platina foil, with soda and nitre, very slight indication of manganese.

With borax, dissolves very slowly; globule transparent, and deep-yellow, hot; colourless, cold; in the reducing flame nearly the same, but colour lighter, and more inclining to green.

With salt of phosphorus, nearly as with borax, except that the action is still slower, the yellow colour, in either flame, lighter, and without any tinge of green. A silica skeleton remains undissolved.

With nitrate of cobalt, dirty-rose colour; the fused edges purple.

From the last result, the grey substance appears to contain a considerable portion of magnesia.

I hope before long to give the analysis of the three preceding minerals.

ARTICLE VII.

Abstract of the Report on M. Rousseau's Memoir respecting a new Method of measuring the Power of Bodies to conduct Electricity. By MM. Ampere and Dulong.*

M. ROUSSEAU, who has been occupied several years in the construction of dry voltaic piles, with the view to discover the circumstances which modify the energy and duration of their action, conceived the idea of employing those instruments to appreciate the different degrees of conducting power of the substances arranged in the class of bad conductors of electricity. For this purpose he has contrived the apparatus we are about to describe. The dry pile, which forms the principal part of it, is made of discs of zinc and tinsel, separated by pieces of parchment, soaked in a mixture of equal parts of oil of poppies, and essence of turpentine; the whole is covered laterally with resin to prevent the contact of the air. The base of the pile

* From the Annales de Chimie,

communicates with the ground. Its upper extremity may be connected by a metallic wire with an insulated vertical pivot, carrying a weakly magnetic needle, balanced horizontally. On a level with the needle, and distant from the pivot, about half the length of the latter, is a metallic ball, also insulated, but communicating with the pile. It is evident that by this arrangement, the electricity accumulated at the upper pole of the pile, is communicated to the needle and the ball, and consequently repulsion ensues, tending to separate the needle, which is moveable, from the ball which is stationary. If we place the pivot and the ball in the magnetic meridian, the needle touches it, and remains at rest as long as the apparatus is not connected with the pile; but the instant the communication is established between them, the needle is repelled, and after some oscillations takes its position of equilibrium, depending on its magnetic power and the energy of the pile. These two quantities remain constant for a considerable time, with the same apparatus, as may be ascertained, by determining the angle which the needle makes with the magnetic meridian, after it has assumed a fixed position, by means of a divided circle adapted to the cage which covers it. A simple conducting needle suspended by a metallic wire of proper diameter and length, might be substituted for the magnetic one; but M. Rousseau's apparatus is much more convenient, and sufficiently sensible for the kind of effect which it is his object to measure.

To use it for ascertaining different degrees of conducting power, it is sufficient to place the substance submitted to experiment in the electrical current, taking care that the thickness which the electricity has to pass through be always equal. If the flow of the quantity of electricity necessary to produce the greatest deviation be not instantaneous, the time required by the needle to assume its fixed position, may be taken as the measure of the degree of the conducting power of the substance employed.

To submit liquids to this kind of examination, M. Rousseau places them in small metallic cups, communicating by their foot with the needle and the ball: he then places in the liquid one of the extremities of the metallic wire, covered with gum lac, that the same surface of metal may always be in contact with the fluid, and measures the duration of the needle's motion from the moment when the communication is established with the pile by the other extremity of the wire.

By submitting the fixed vegetable oils employed in the arts and in domestic economy to this kind of proof, M. Rousseau has established a very singular fact, which may be useful in commerce; it is that olive oil possesses a very inferior degree of conducting power to that of all the other vegetable or animal oils, which nevertheless present, in all their physical proper